APPLICATIONS:
- Precise radiation detection, radiation spectroscopy
TECHNOLOGY DESCRIPTION:
SPECTRE-ML is a machine learning-based software for finding optimal clusters of radiation detector segments (pixels/voxels) to enhance spectroscopic performance of detectors. By learning detector-specific trends—such as dead layers, edge effects, and gain shifts—it outperforms traditional methods like uniform depth clustering. The platform streamlines the entire workflow, from data preprocessing and ML algorithm execution to output evaluation and visualization. Designed for highly-segmented CdZnTe (CZT) detectors in International Atomic Energy Agency (IAEA) non-destructive assay safeguards tasks, SPECTRE-ML ensures precise nuclear material verification. Its adaptable framework also supports extension to other segmented detectors, making it a versatile tool for next-generation radiation spectroscopy.
PRINCIPAL INVESTIGATORS:
- Jayson Vavrek
STATUS: Copyrighted
OPPORTUNITIES: Available for licensing or collaborative research.